ContainersNVIDIADeepSeek-R1NVIDIALlama-3.1-Nemotron-70B-InstructNVIDIAPyTorch
CollectionsNVIDIAOmniverse Kit (FB)NVIDIADeepStream SDKNVIDIAOmniverse Kit App Streaming
ModelsNVIDIAStyleGAN3 pretrained modelsNVIDIAPeopleNetNVIDIATrafficCamNet
ResourcesNVIDIARiva Skills Quick Start
Helm ChartsNVIDIAGPU OperatorNVIDIANVIDIA NIM Operator
NVIDIA
NVIDIA
PyTorch
Container
NVIDIA
NVIDIA
PyTorch

PyTorch is a GPU accelerated tensor computational framework. Functionality can be extended with common Python libraries such as NumPy and SciPy. Automatic differentiation is done with a tape-based system at the functional and neural network layer levels.

NVIDIA AI Enterprise Supported
LayerLabelCreated
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4LABEL
com.nvidia.build.ref=40d6cf911ff7816d89ec212ae4ba065395be1d07
11/15/2024 2:25 AM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ARG
NVIDIA_BUILD_REF=40d6cf911ff7816d89ec212ae4ba065395be1d07
11/15/2024 2:25 AM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4LABEL
com.nvidia.build.id=122402535
11/15/2024 2:25 AM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ENV
NVIDIA_BUILD_ID=122402535
11/15/2024 2:25 AM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ARG
NVIDIA_BUILD_ID=122402535
11/15/2024 2:25 AM UTC
sha256:1f630473117188fe348ebdcf0da5e93138275af14007f15baf79967a365e647aCOPY
entrypoint.d/ /opt/nvidia/entrypoint.d/
11/15/2024 2:25 AM UTC
sha256:9974f1524ebf95ed7a0c34b41352f4b1c470afc3c2b85d06baae95c63351bd8aRUN
NVIDIA_PYTORCH_VERSION=24.11 PYTORCH_BUILD_VERSION=2.6.0a0+df5bbc0 NVFUSER_BUILD_VERSION=0d33366 TARGETARCH=arm64 PYVER=3.12 PYVER_MAJMIN=312 L4T=1 PYVER=3.12 ln -sf ${_CUDA_COMPAT_PATH}/lib.real ${_CUDA_COMPAT_PATH}/lib &&
  echo ${_CUDA_COMPAT_PATH}/lib > /etc/ld.so.conf.d/00-cuda-compat.conf &&
  ldconfig &&
  rm -f ${_CUDA_COMPAT_PATH}/lib
11/15/2024 2:25 AM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ENV
CUDA_MODULE_LOADING=LAZY
11/15/2024 2:25 AM UTC
sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4ENV
TORCH_CUDNN_V8_API_ENABLED=1
11/15/2024 2:25 AM UTC
sha256:e6637e27ac4fd8a36683757c1b1d6a24d42aeaeff1f88a281e07d6ab08ec0c80RUN
NVIDIA_PYTORCH_VERSION=24.11 PYTORCH_BUILD_VERSION=2.6.0a0+df5bbc0 NVFUSER_BUILD_VERSION=0d33366 TARGETARCH=arm64 PYVER=3.12 PYVER_MAJMIN=312 L4T=1 PYVER=3.12 if [ "${L4T}" = "1" ]; then echo "Not installing Transformer Engine in iGPU container until Version variable is set"; else NVTE_BUILD_THREADS_PER_JOB=8 pip install --no-cache-dir --no-build-isolation git+https://github.com/NVIDIA/TransformerEngine.git@release_v${TRANSFORMER_ENGINE_VERSION}; fi
11/15/2024 2:25 AM UTC
...